SkillRise: Agentic Reinforcement Learning for Cross-Task Skill Evolution

• URL: http://arxiv.org/abs/2607.26784v1
• Date: Wed, 29 Jul 2026 11:26:33 GMT
• Title: SkillRise: Agentic Reinforcement Learning for Cross-Task Skill Evolution
• Abstract summary: Large language model agents often encounter related yet distinct tasks that share reusable solution patterns.<n>We introduce SkillRise, a unified reinforcement learning framework for learning skills across tasks.
• Score: 39.74364990208475
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language model agents often encounter related yet distinct tasks that share reusable solution patterns. Yet standard agentic reinforcement learning treats tasks as independent episodes, while existing approaches to skill learning either focus on repeated attempts of one task or use pipelines with multiple stages that entangle extraction, retrieval, and execution. We introduce SkillRise, a unified reinforcement learning framework for learning skills across tasks. SkillRise organizes related instances into progressively challenging sequences and uses a single policy to alternate between task solving and curating an evolving skill document passed directly to the next task. Decoupled credit assignment across tasks supervises solving with the current task outcome and curation with discounted downstream outcomes. Experiments on ALFWorld, WebShop, and ScienceWorld show that SkillRise achieves the strongest Pass@1 performance among the compared methods, with gains over the strongest baseline ranging from 2.3 to 8.5 percentage points. Although trained across distinct tasks, its learned curation policy remains effective for repeated attempts on the same task. Further analysis reveals scaling at test time across tasks: performance improves with longer sequences of related tasks even when each task is attempted only once. This trend suggests that SkillRise reuses transferable skills across tasks rather than benefiting from repeated sampling of the same task. SkillRise further retains strong performance while substantially reducing the runtime overhead of skill learning pipelines with multiple stages. Together, these results provide a simple and efficient training paradigm for LLM agents to extract, refine, and reuse transferable skills across tasks.

Related papers

• SkillOS: Learning Skill Curation for Self-Evolving Agents [67.94374107466957]
  We propose SkillOS, an experience-driven RL training recipe for learning skill curation in self-evolving agents.<n>SkillOS pairs a frozen agent executor that retrieves and applies skills with a trainable skill curator that updates an external SkillRepo from accumulated experience.<n>We show that SkillOS consistently outperforms memory-free and strong memory-based baselines in both effectiveness and efficiency.
  arXiv  Detail & Related papers  (2026-05-07T17:31:50Z)
• Minimax Forward and Backward Learning of Evolving Tasks with Performance Guarantees [6.008132390640294]
  The incremental learning of a growing sequence of tasks holds promise to enable accurate classification. This paper presents incremental minimax risk classifiers (IMRCs) that effectively exploit forward and backward learning. IMRCs can result in a significant performance improvement, especially for reduced sample sizes.
  arXiv  Detail & Related papers  (2023-10-24T16:21:41Z)
• Multitask Learning with No Regret: from Improved Confidence Bounds to Active Learning [79.07658065326592]
  Quantifying uncertainty in the estimated tasks is of pivotal importance for many downstream applications, such as online or active learning. We provide novel multitask confidence intervals in the challenging setting when neither the similarity between tasks nor the tasks' features are available to the learner. We propose a novel online learning algorithm that achieves such improved regret without knowing this parameter in advance.
  arXiv  Detail & Related papers  (2023-08-03T13:08:09Z)
• Active Task Randomization: Learning Robust Skills via Unsupervised Generation of Diverse and Feasible Tasks [37.73239471412444]
  We introduce Active Task Randomization (ATR), an approach that learns robust skills through the unsupervised generation of training tasks. ATR selects suitable tasks, which consist of an initial environment state and manipulation goal, for learning robust skills by balancing the diversity and feasibility of the tasks. We demonstrate that the learned skills can be composed by a task planner to solve unseen sequential manipulation problems based on visual inputs.
  arXiv  Detail & Related papers  (2022-11-11T11:24:55Z)
• Is Multi-Task Learning an Upper Bound for Continual Learning? [26.729088618251282]
  This paper proposes a novel continual self-supervised learning setting, where each task corresponds to learning an invariant representation for a specific class of data augmentations. We show that continual learning often beats multi-task learning on various benchmark datasets, including MNIST, CIFAR-10, and CIFAR-100.
  arXiv  Detail & Related papers  (2022-10-26T15:45:11Z)
• Task Compass: Scaling Multi-task Pre-training with Task Prefix [122.49242976184617]
  Existing studies show that multi-task learning with large-scale supervised tasks suffers from negative effects across tasks. We propose a task prefix guided multi-task pre-training framework to explore the relationships among tasks. Our model can not only serve as the strong foundation backbone for a wide range of tasks but also be feasible as a probing tool for analyzing task relationships.
  arXiv  Detail & Related papers  (2022-10-12T15:02:04Z)
• Towards More Generalizable One-shot Visual Imitation Learning [81.09074706236858]
  A general-purpose robot should be able to master a wide range of tasks and quickly learn a novel one by leveraging past experiences. One-shot imitation learning (OSIL) approaches this goal by training an agent with (pairs of) expert demonstrations. We push for a higher level of generalization ability by investigating a more ambitious multi-task setup.
  arXiv  Detail & Related papers  (2021-10-26T05:49:46Z)
• Efficiently Identifying Task Groupings for Multi-Task Learning [55.80489920205404]
  Multi-task learning can leverage information learned by one task to benefit the training of other tasks. We suggest an approach to select which tasks should train together in multi-task learning models. Our method determines task groupings in a single training run by co-training all tasks together and quantifying the effect to which one task's gradient would affect another task's loss.
  arXiv  Detail & Related papers  (2021-09-10T02:01:43Z)
• CrossFit: A Few-shot Learning Challenge for Cross-task Generalization in NLP [38.40614678878222]
  We introduce CrossFit, a task setup for studying cross-task few-shot learning ability. We present NLP Few-shot Gym, a repository of 160 few-shot NLP tasks. Our empirical analysis reveals that the few-shot learning ability on unseen tasks can be improved via an upstream learning stage.
  arXiv  Detail & Related papers  (2021-04-18T12:14:46Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.